Thomas Wagenaar

If you're using a LSTM, you always set the output of a training case to be the input of the next training case. So if the value of my stocks...

> since the values are normalized, the network learns changes in percentage and not value, right? Internally the network 'knows' nothing. You can input the values in whatever form you...

Have you tried looking at existing projects? e.g. https://wagenaartje.github.io/neataptic/articles/agario/

I completely understand. The way it works is quite different to the original NEAT algorithm. I actually call it the 'Instict' algorithm ([more about that can be read there](https://medium.com/towards-data-science/neuro-evolution-on-steroids-82bd14ddc2f6)). I'm...

@pjbaron It's a long time since I have read the paper, but I do not remember how innovation numbers maintain diversity. What I do recall is that innovation numbers are...

Thanks! Again, my knowledge on the library / algorithm is a bit stale, so there might be some incorrect facts in the piece below. I completely forgot to mention, **the...

Thanks for all the kind comments, and I appreciate that people have created projects that continue with Neataptic. Without promising anything, I would like to share that I am working...

It is a full rewrite @dan-ryan. The first release will add no new features, but it is purely based on speed performance. If this works out, the project will be...

Please turn your code into a code block (See "Syntax Highlighting" [here](https://guides.github.com/features/mastering-markdown/)). I'm glad to help, but I need more info on what exactly you want to improve and what...

Which error is occurring?